Therapeutic value of patent foramen ovale closure for drug‐resistant epilepsy: A case series report

Abstract

Objective

Closure surgery of patent foramen ovale (PFO) has been found to effectively control cryptogenic stroke and migraine, but it is uncertain whether PFO closure could also alleviate epileptic seizures. This study aims to observe the therapeutic effect of PFO closure on epileptic seizures.

Methods

Since July 11th, 2017, in the neurology department of West China Hospital, Sichuan University, Chengdu, we have been regularly monitoring patients with epilepsy who have undergone PFO closure. The patient's clinical information, such as frequency, duration, and severity of seizures, before and after surgery was recorded in detail as well as postoperative safety events.

Results

Of the 31 epilepsy patients who confirmed PFO observed (27 cases were drug‐resistant epilepsy, 87.10%), average age of surgery was 23.74 years, and 12 cases were female (38.71%). After one‐year follow‐up, 26 patients (83.87%) achieved remission of seizure frequency, and 22 of whom (70.97%) experienced a remission of more than 50%. Additionally, compared to before surgery, 22 cases (70.97%) reported a decrease in the average seizure duration, and 20 cases (64.52%) reported a reduction in seizure severity. In the seizure indicators of frequency, average duration and severity, significant differences were identified between preoperative and postoperative comparisons with all test p values were <0.05. Furthermore, no serious safety events were reported except for one patient who briefly reported chest pain, and all patients expressed effective PFO closure.

Significance

The PFO closure has been shown for the first time to result in a significant reduction in the frequency, duration, and severity of seizures. Patients with drug‐resistant epilepsy and PFO with a large shunt are ideal candidates for undergoing PFO closure.

Plain Language Summary

Since PFO closure was found to have a good therapeutic effect on cryptogenic stroke and migraine, it has become a credible complementary therapy for the treatment of neurological diseases, and drug‐resistant epilepsy with PFO is expected to become the next target disease that PFO closure could significantly improve.

Key points.

• Patent foramen ovale (PFO) closure could relieve seizures symptoms for patients with epilepsy.

• Patients with drug‐resistant epilepsy and PFO with a large right‐to‐left shunt volume are suitable for undergoing PFO closure.

• PFO closure is expected to be a potential peripheral organ approach for the treatment of epilepsy.

1. INTRODUCTION

Congenital heart disease (CHD) is a common birth defect that affects the normal development of the heart and nervous system in newborns. ¹ Due to the tremendous advances in surgical methods and intensive care, short‐term survival rates of complex CHD patients have considerably improved, but long‐term exposure to simple CHD could induce neurodevelopmental harm, which has become one of the focal areas of current research. ² , ³ For instance, Michelle et al. investigated 15 222 CHD patients in Danish hospital using population registries, and revealing that the hazard ratio of epilepsy in CHD patients during long‐term observation is 2.3–3.7 times than that of general population. ⁴ The studies conducted by Billett et al. and Zelano et al. were also reported this phenomenon. ⁵ , ⁶

In the observation of these simple CHD patients, certain researches indicate that the surgery for CHD might raises the risk of perioperative seizures, ⁷ but some studies suggest that simple CHD patients who received surgical intervention could reduce the long‐term incidence of epilepsy. ⁴ Currently, there is still ongoing debate about the impact of simple CHD surgery on the likelihood of developing epilepsy, which has sparked our interest and contemplation.

Studies on patent foramen ovale (PFO), a common congenital heart disease that affects 20%–34% of general population, ⁸ have demonstrated that closuring PFO could greatly reduce the risk of migraine and cryptogenic stroke, especially in patients with a large shunt volume. ⁹ , ¹⁰ , ¹¹ However, epilepsy is a neurological disorder closely related to migraine, there are no studies that have explored the association between epilepsy and PFO. In 2018, among 339 epilepsy patients, our team first discovered that patients with moderate to large right‐to‐left shunt (RLS) had a higher prevalence of migraines (39% vs. 18%, OR = 2.90, with 95% CI = 1.41–5.98, p‐value = 0.003). ¹² Furthermore, in 2023, our team also reported that PFO was more common in epilepsy patients than in the general population (39% in epilepsy patients vs. 24.25% in the general population). ¹³

From the view of pathological mechanism, causing the PFO permits blood to bypass the pulmonary circulation and immediately enter the systemic circulation, various substances may therefore be carried into the brain, including air, choline compounds, glutamine, inflammatory mediators, microemboli, tiny molecules, and a portion of deoxygenated venous blood. ¹⁴ , ¹⁵ , ¹⁶ Our exploration found that migraine and epilepsy display consistent blood proteomic changes but exhibit significant differences at the metabolic level. Furthermore, the presence of patent foramen ovale (PFO) exacerbates molecular alterations in both conditions, including increased keratin, elevated bile acid metabolism, disruptions in immune response, and neurotransmitter imbalances.

For epilepsy patients, although most patients could get effective seizure control through antiseizure medications (ASMs), there are still about one‐third of patients who cannot achieve seizure freedom through ASMs, which makes these patients have been facing huge psychological pressure and serious disease burden. Since 2017, our team observed that among the patients with drug‐resistant epilepsy who were routinely monitored in our hospital, three drug‐resistant patients accompanied with migraine not only relived the symptoms of migraine after PFO closure surgery but also achieved the remission of seizure frequency.

Based on the evidences presented, we conducted a prospective observational study in West China Hospital of Sichuan University. Considering that patients with non‐drug‐resistant epilepsy could control their seizures with antiseizure drugs, we inclined to advocate PFO closure surgery for patients with drug‐resistant epilepsy and PFO. After that, over a lengthy follow‐up, we recorded the features of their seizures. In patients with drug‐resistant epilepsy and a sizable right‐to‐left shunt, our study seeks to determine whether PFO closure can reduce epileptic seizures.

2. METHODS

2.1. Patient registration and statement

The neurology department at our hospital registered this prospective case‐series study on July 10th, 2017, and the West China Hospital of Sichuan University Ethics Committee authorized it (ChiCTR‐OOC‐17011935). It was advised that patients with drug‐resistant epilepsy have cardiac ultrasonography under informed consent in order to determine whether or not they had a significant right‐to‐left shunt. Patients who met the eligibility requirements might choose to have their PFOs closed, and those who did were monitored for a minimum of 1 year after the treatment. The hospital's ultrasonography, cardiology, and neurology departments independently performed imaging examinations, the PFO closure surgery, and the process of diagnosing and monitoring epilepsy for all patients. The planning and design of this study involved neither the public nor the patients.

2.2. Cases definition and PFO diagnosis

Epilepsy was confirmed using the 2014 International League Against Epilepsy (ILAE) standards. ¹⁷ The etiological causes and the epileptic syndromes were diagnosed according to the diagnostic criteria of epilepsy etiology classification of ILAE (2017 version) and the classification and definition of epilepsy syndromes with onset at a variable age (2022 version) ¹⁸ , ¹⁹ , ²⁰ ; of these, the etiological causes could be classified as six categories: structural, genetic, infectious, metabolic, immune, and unknown; and the description of epileptic syndromes refers to a cluster of features incorporating seizure types, electroencephalography (EEG), and magnetic resonance imaging (MRI) that tend to occur together.

The identified cases of drug‐resistant epilepsy based on the ILAE definition of drug‐resistant epilepsy (2010 version), which required the epilepsy patient to comply with the failure of adequate trials of two tolerated, appropriately chosen and used antiepileptic drug schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom. In our hospital, we confirmed patients with drug‐resistant epilepsy who had failed to effectively control their seizures after taking two ASMs for at least more than 1 year, the diagnostic process was determined jointly by two experienced neurologists, and the diagnostic process also involved assessing seizure frequency and symptoms, as well as reviewing EEG and MRI results. ²¹ In this study, we selected as many patients as possible who were diagnosed with drug‐resistant epilepsy; however, several patients who had not yet been identified with a diagnosis of drug‐resistant epilepsy were also included in this study because of their strong personal desire to participate.

The diagnosis of PFO was conformed by the American Society of Echocardiography and Society for Cardiac Angiography and Intervention (ASE and SCAI, 2015 version). ²² Patients who underwent PFO closure were required to have at least a secondary or higher grade of RLS volume, which was detected using transthoracic echocardiography (TTE) and the observation of shunting microbubbles following contrast injection. ²³ For semiquantitative assessment, the RLS volume detected by TTE is categorized into four grades based on the number of microvesicles detected per frame in the left atrium: grade 0 = no microbubbles present, grade I = 1–10 microbubbles, grade II = 11–30 microbubbles, and grade III = over 30 microbubbles, or left atrium almost filled with microbubbles, or left atrial opacity. ²⁴ If the RLS volume of PFO patients exceeded grade I, individuals were advised to undergo transesophageal echocardiography (TEE). All relevant examination procedures are detailed in Appendix S1.

2.3. PFO closure process and medication strategies

After conducting TTE and TEE assessments, an experienced interventional cardiologist (Y.C.) evaluated the clinical indications of transcatheter PFO closure based on a consensus among Chinese experts. ²⁵ The interventional procedure was than performed via femoral vein under local anesthesia. All participants were required strictly to complete a postoperative antiplatelet therapy regimen, which included two stages of daily medication: 75 mg of clopidogrel and 100 mg of aspirin daily for the first month, followed by 100 mg of aspirin daily from the second to sixth months. If there were any significant adverse effects from the antiplatelet therapy, the drugs were discontinued. Chest radiographs and TTE were the primary methods used for post‐procedure follow‐up to ensure the devices were properly positioned.

Following PFO closure, all subjects were suggested to continue to maintain the same ASMs administration strategy as before surgery. During the postoperative follow‐up period, the ASMs administration will be changed accordingly when subjects were found have an exacerbation of seizures, and we recorded the changes of patient's ASMs regimen during the follow‐up period in detail.

2.4. Variables collection and statistical analysis

Healthcare providers collected the detailed information of demographics, seizure history and symptoms (including the onset age, course of epilepsy, seizure types, whether accompanied by aura or migraine, etc.), etiological causes (including types of structural, genetic, infectious, metabolic, immune and unknown), duration and types of ASMs before surgery the ASMs duration represents the length of time from the time at which the patients started the anti‐seizure medications until 1 week before surgery, and the types of ASMs including Valproate (VPA), Levetiracetam (LEV), Lamotrigine (LTG), Topiramate (TPM), Oxcarbazepine (OXC), Phenobarbital (PB), Carbamazepine (CBZ), as well the diagnosis of EEG and MRI during the initial consultation. For patients included in the analysis, subjective assessments of seizure symptoms and clinical information were regularly recorded during postoperative clinic visits and telephone calls on the 1st, 3rd, 6th and 12th months following the surgery. Following PFO closure, all subjects were suggested to continue to maintain the same ASMs administration strategy as before surgery, and we recorded the changes in the patient's ASMs regimen during the follow‐up period.

The measures used to evaluate the effectiveness of the surgery in managing epileptic seizures included the frequency, duration, and severity of seizures, which were assessed using the shortened Ictal/Post‐ictal subscale of the Liverpool Seizure Severity Scale. ²⁶ This scale includes dimensions such as lip smacking or fidgeting, falling, headache and sleepiness after recovery, incontinence, tongue biting and injury during an attack. Safety evaluation took into account postoperative adverse effects, such as atrial fibrillation, paroxysmal supraventricular tachycardia, thrombosis, aortic dissection, occluder displacement, and any other related issues. ¹¹

The primary outcome was the reduction percentage of the seizures frequency in the first year after PFO closure procedure compared to the year before the surgery. The study defined remission and greater remission according to whether the reduction percentage was above 0% and above 50%, respectively. Paired Shapiro–Wilk test was used to compare the difference of seizure frequency before and after PFO surgery, and Chi‐square analysis and Wilcoxon rank‐sum test were used to compare each clinical indicators between patients who achieved remission and who did not. The statistical analysis was carried out using R version 4.3.1 (R Foundation for Statistical Computing, using packages such as mice, stats, and tidyverse). All p values were two‐tailed, and statistical significance was set at 0.05 or 0.1.

3. RESULTS

3.1. Patients description

Since July 10, 2017, a total of 35 patients voluntarily received PFO closure in our hospital, and 4 cases among them were excluded from the observational period. The exclusions were due to one patient's failure to have the PFO occluder installed, one patient's ineffective control of postoperative RLS volume, and two patients not receiving long‐term postoperative follow‐up. Of the 31 patients who were remained, 27 cases were confirmed as drug‐resistant epilepsy with the proportion of 87.10%, and the demographic and clinical details are presented in Table 1. The average age of the subjects at the time of PFO closure was 23.74, ranging from 6 to 50 years old, with 12 cases (38.71%) being female. The average onset age of seizure was 16.26 years old, with an average epilepsy course of 7.71 years, 19 cases (61.29%) had a third‐grade RLS volume, 7 cases (22.58%) had seizures with aura and 8 cases (25.81%) were also diagnosed with migraine.

TABLE 1.

Demographic characteristics and clinical information of patients with epilepsy and PFO at baseline.

Number	Sex	Age (years)	Onset age (years)	BMI (kg/m2)	EP course (years)	Aura	Migraine	RLS grade	Seizure type	Etiological causes	EEG	MRI	Drug‐resistant epilepsy
1	Female	15	12	22.58	4	No	No	II	Generalized	Unknown	Abnormal	Normal	Yes
2	Male	36	28	19.36	8	No	No	III	Focal	Unknown	Abnormal	Normal	Yes
3	Female	50	49	21.64	1	No	No	III	Unknown	Unknown	Normal	Normal	No
4	Female	42	22	25.39	21	No	Yes	III	Focal	Unknown	Normal	Minor ischemic focus	Yes
5	Male	29	12	22.86	17	Yes	Yes	II	Generalized	Unknown	Abnormal	Normal	Yes
6	Male	20	11	20.31	9	No	No	III	Unknown	Unknown	Normal	Normal	Yes
7	Male	23	20	18.93	3	Yes	Yes	III	Focal	Unknown	Normal	Normal	Yes
8	Male	21	17	22.66	4	No	No	III	Focal	Structural	Abnormal	Structural	Yes
9	Male	21	15	25.25	6	No	No	III	Unknown	Structural	Normal	Hippocampal high signal	Yes
10	Male	31	15	28.34	17	No	No	III	Focal	Unknown	Abnormal	Normal	Yes
11	Female	22	21	22.89	1	No	No	III	Generalized	Structural	Normal	Cortical dysplasia	Yes
12	Female	47	45	19.56	2	No	No	II	Focal	Unknown	Abnormal	Normal	Yes
13	Male	35	35	19.72	1	No	No	II	Unknown	Unknown	Normal	Normal	No
14	Female	14	12	19.72	2	No	Yes	II	Unknown	Unknown	Normal	Normal	Yes
15	Female	22	10	22.22	12	Yes	Yes	III	Focal	Unknown	Abnormal	Normal	Yes
16	Male	33	15	22.49	18	No	Yes	II	Unknown	Unknown	Normal	Normal	Yes
17	Male	24	19	20.98	6	Yes	No	III	Focal	Unknown	Abnormal	Normal	Yes
18	Female	29	29	20.31	1	No	No	III	Unknown	Genetic	Normal	Normal	No
19	Male	13	12	16.36	1	No	No	III	Focal	Unknown	Abnormal	Normal	Yes
20	Female	12	3	18.60	9	Yes	No	III	Focal	Unknown	Normal	Normal	Yes
21	Male	22	12	28.58	11	No	No	III	Generalized	Unknown	Normal	Normal	Yes
22	Female	21	16	17.90	5	No	No	II	Focal	Unknown	Normal	Normal	Yes
23	Male	15	9	21.48	6	No	No	II	Focal	Structural	Normal	Hippocampal sclerosis	Yes
24	Male	23	12	22.05	11	Yes	No	III	Focal	Unknown	Normal	Normal	Yes
25	Male	17	9	20.76	8	No	No	II	Focal	Structural	Normal	Arachnoid cyst	Yes
26	Male	15	3	19.49	12	No	Yes	II	Focal	Structural	Normal	Cortical dysplasia	Yes
27	Male	6	0	18.00	6	No	No	II	Generalized	Unknown	Abnormal	Normal	Yes
28	Female	31	30	20.70	1	No	No	III	Focal	Unknown	Abnormal	Malacia in parietal lobe	Yes
29	Female	26	1	20.20	25	Yes	No	II	Focal	Infectious	Abnormal	Normal	Yes
30	Male	9	8	31.53	1	No	Yes	III	Focal	Unknown	Abnormal	Normal	No
31	Male	12	2	16.65	10	No	No	III	Unknown	Infectious	Normal	Normal	Yes

Note: Variables classification: RLS grade including N (no occurrence of micro‐bubbles), I (1–10 micro‐bubbles), II (11–30 micro‐bubbles) and III (over 30 micro‐bubbles); sex including female and male; seizure types including focal onset, generalized onset and unknown onset; etiological causes including structural, genetic, infectious, metabolic, immune and unknown; aura, migraine and drug‐resistant epilepsy including No and Yes. Variables definition: Abnormal EEG including sharp, spike and slow waves found in the frontal, parietal, temporal, and occipital lobes of the brain and in the hippocampus or globally; abnormal MRI including obvious bleeding foci, arachnoid cysts or hippocampal sclerosis were found in the frontal, parietal, temporal and occipital lobes of the brain and the hippocampus. Variables units: units of age, onset age and EP course before surgery are years; unit of BMI is kg/m².

Abbreviations: BMI, body mass index; EEG, electroencephalography; EP, epilepsy; MRI, magnetic resonance imaging; RLS, right‐to‐left shunt.

From the view of seizure types, 18 cases were confirmed as focal onset (58.06%), 5 cases were generalized onset (16.13%) and 8 cases were unknown onset (25.81%); and from the insight of etiological causes, 22 cases were classified as unknown etiology (70.97%), and 9 cases were classified as structural, genetic or infectious etiologies (29.03%); the detailed description of seizure performance for each patients is shown in Table 2. Among these subjects, the average duration of taken ASMs before surgery was 4.84 years, and 15 cases (48.39%) were taking more than one type of ASMs before surgery. All participants were followed up for at least 1 year and 9 cases were recorded the changes of ASMs regimen during the follow‐up period, as details shown in Table 3.

TABLE 2.

Description of preoperative seizure performance and anti‐seizure medication regimens for patients with epilepsy and PFO.

Number	Seizure performance	ASMs duration (years)	ASMs types	ASMs regimens
1	Feeling dizziness, upper limb myoclonic seizures	3.33	3	VPA 500 mg, LEV 500 mg, LTG 50 mg
2	Sensation of cold and ictal piloerection with or without progressing to clonic–tonic state	8.00	2	TPM 150 mg, VPA 1000 mg
3	Generalized tonic–clonic seizures during sleep	0.33	1	LEV 500 mg
4	Hearing a ringing or buzzing sound, generalized tonic–clonic seizures	10.00	1	LTG 50 mg
5	Generalized tonic–clonic seizures during sleep	10.00	2	OXC 600 mg, LEV 1000 mg
6	Generalized tonic–clonic seizures	5.00	1	VPA 500 mg
7	Sensation of heat and deja vu with or without progressing to tonic–clonic seizures	1.33	2	LEV 500 mg, OXC 150 mg
8	Automotor seizures after deja vu	4.00	1	OXC 1200 mg
9	Generalized tonic–clonic seizures during sleep	1.33	2	VPA 500 mg, OXC 300 mg
10	Feeling of numbness of left arm and palpitation, and then ictal unilateral dystonia progress to tonic–clonic seizures	16.25	4	LEV 750 mg, LTG 100 mg, TPM 50 mg, VPA 250 mg
11	Giggle, stare blankly with grabbling, generalized tonic–clonic seizures	1.00	2	OXC 300 mg, LEV 500 mg
12	Generalized tonic–clonic seizures	1.00	1	VPA 500 mg
13	Staring spells, chewing, lip smacking, and then progress to generalized tonic–clonic seizures	0.08	1	OXC 600 mg
14	Stare blankly, generalized tonic–clonic seizures	1.75	3	LEV 500 mg, OXC 300 mg, TPM 25 mg
15	Feeling of disgusting and then progress to tonic–clonic seizures	1.00	1	OXC 900 mg
16	Generalized tonic–clonic seizures during sleep	2.17	2	OXC 600 mg, VPA 1000 mg
17	Feeling of fear, palpitation, with or without tonic–clonic seizures	0.00	0	Abandoned to take ASMs because of poor efficacy
18	Tonic seizures after a moaning sound	0.08	1	VPA 500 mg
19	Generalized tonic–clonic seizures	1.00	2	OXC 750 mg, LEV 1500 mg
20	Generalized tonic–clonic seizures during sleep	7.00	1	LEV 500 mg
21	Generalized tonic–clonic seizures	10.50	2	PB 60 mg, VPA 500 mg
22	Feeling of numbness on face, palpitation, and then loss of consciousness with or without tonic seizures	1.17	1	LEV 1000 mg
23	Stare blankly, turn to right, generalized tonic–clonic seizures	6.00	1	OXC 600 mg
24	Flashes of light and deja vu, then progressing to tonic–clonic seizures	11.00	2	VPA 500 mg, OXC 300 mg
25	Generalized tonic–clonic seizures	2.00	2	VPA 500 mg, LTG 25 mg
26	Right upper limb myoclonic seizures	12.00	1	OXC 600 mg
27	Generalized tonic–clonic seizures during sleep	1.00	1	VPA 750 mg
28	Right leg feeling uncomfortable, generalized tonic–clonic seizures	1.00	1	LEV 1500 mg
29	Palpitation aura, generalized tonic–clonic seizures	20.00	2	LEV 1000 mg, CBZ 600 mg
30	Generalized tonic–clonic seizures during sleep	0.67	1	LEV 500 mg
31	Feeling dizziness, turn to right, generalized tonic–clonic seizures	10.00	3	VPA 400 mg, LEV 1000 mg, TPM 50 mg

Note: Variables definition: ASMs duration before surgery represents the length of time from the time at which the patients started the anti‐seizure medications until 1 week before surgery; ASMs types before surgery represents the number of ASM categories taken by patients in the week before surgery, including Valproate (VPA), Levetiracetam (LEV), Lamotrigine (LTG), Topiramate (TPM), Oxcarbazepine (OXC), Phenobarbital (PB), Carbamazepine (CBZ). Variables units: Units of ASMs duration before surgery is years.

Abbreviation: ASMs, anti‐seizure medications.

TABLE 3.

Description of postoperative anti‐seizure and anti‐platelet medication regimens and safety events for patients with epilepsy and PFO.

Number	ASMs types	ASMs regimens	Whether add medicine	Clopidogrel administration (months)	Aspirin administration (months)	Safety events	Follow‐up
1	3	VPA 500 mg, LEV 500 mg, LTG 50 mg	Sustained	6	12	No	3.56
2	2	TPM 75 mg, OXC 450 mg	TPM changed as OXC	6	12	No	4.93
3	1	LEV 500 mg	Sustained	6	12	No	2.85
4	1	LTG 50 mg	Sustained	6	12	No	3.81
5	2	OXC 600 mg, LEV 1000 mg	Sustained	6	12	No	3.87
6	1	VPA 500 mg	Sustained	6	12	No	2.95
7	2	LEV 500 mg, OXC 150 mg	Sustained	6	12	No	2.72
8	1	OXC 1200 mg	Sustained	6	12	No	5.29
9	2	VPA 500 mg, LEV 500 mg	OXC changed as LEV	6	12	No	3.41
10	4	LEV 1000 mg, LTG 100 mg, TPM 50 mg, VPA 250 mg	Dosage of LEV added to 1000 mg/day	6	12	No	2.57
11	2	OXC 300 mg, LEV 500 mg	Sustained	6	12	No	2.96
12	1	VPA 500 mg	Sustained	6	12	No	5.10
13	1	OXC 600 mg	Sustained	6	12	No	5.51
14	3	LEV 500 mg, OXC 300 mg, TPM 25 mg	Sustained	6	12	No	4.04
15	1	OXC 1200 mg	Dosage of OXC added to 1200 mg/day	6	12	No	5.29
16	3	OXC 600 mg, VPA 1000 mg, TPM 125 mg	6mth added TPM	6	12	No	5.31
17	1	VPA 500 mg	1mth added VPA	3	3	No	3.76
18	1	VPA 500 mg	Sustained	6	12	No	3.88
19	3	OXC 750 mg, LEV 500 mg, VPA 750 mg	12mth added VPA	6	12	No	3.81
20	1	LEV 500 mg	Sustained	6	12	No	2.97
21	2	LTG 150 mg, VPA 500 mg	PB changed as LTG	6	12	No	3.07
22	1	LEV 1000 mg	Sustained	6	12	No	3.08
23	3	OXC 600 mg, LTG 50 mg, VPA 500 mg	8mth added LTG and VPA	6	12	No	3.10
24	2	VPA 500 mg, OXC 300 mg	Sustained	6	12	Chest pain	3.05
25	2	VPA 500 mg, LTG 25 mg	Sustained	6	12	No	3.07
26	1	OXC 600 mg	Sustained	6	12	No	5.41
27	1	VPA 750 mg	Sustained	6	12	No	5.41
28	1	LEV 1500 mg	Sustained	6	12	No	5.35
29	2	LEV 1000 mg, CBZ 600 mg	Sustained	6	12	No	5.33
30	1	LEV 500 mg	Sustained	6	12	No	3.95
31	3	VPA 400 mg, LEV 1000 mg, TPM 50 mg	Sustained	6	12	No	3.91

Note: Variables definition: ASMs types after surgery represents the number of ASM categories taken by patients in the follow‐up period after PFO closure, including Valproate (VPA), Levetiracetam (LEV), Lamotrigine (LTG), Topiramate (TPM), Oxcarbazepine (OXC), Phenobarbital (PB), Carbamazepine (CBZ). Variables units: Units of anti‐platelet medication after surgery and Follow‐up period are years.

Abbreviations: ASMs, anti‐seizure medications; mth, months.

3.2. Seizure outcomes

The comparison of epileptic seizure characteristics before and after operation is presented in Table 4. The results of the paired Shapiro–Wilk test indicated that the postoperative seizure frequency was significantly lower than pre‐operation (p < 0.01), the postoperative seizure duration was also significantly shorter than pre‐operation (p < 0.001), and the postoperative seizure severity was significantly less severe than pre‐operation (p < 0.001). In terms of the primary outcome, 26 cases (83.87%) of the total enrolled patients achieved remission, with 22 cases (70.97%) experiencing remission >50%. Regarding the secondary outcome, 22 cases (70.97%) exhibited a reduction in the average duration of seizure after PFO closure, and 20 cases (64.52%) expressed a remission in seizure severity out of 31 cases.

TABLE 4.

The seizure characteristics of patients with epilepsy and PFO before and after PFO closure, as well as the efficacy of remission for seizure characteristics.

Number	Preoperative epileptic seizures			Postoperative epileptic seizures			Frequency remission	Frequency remission over 50%	Duration remission	Severity remission
	Frequency (per year)	Average duration (min)	Severity	Frequency (per year)	Average duration (min)	Severity
1	1	0.50	1	0	0.00	0	Achieved	Achieved	Achieved	Achieved
2	1460	0.17	0	540	0.17	0	Achieved	Achieved	None	None
3	3	2.00	0	0	0.00	0	Achieved	Achieved	Achieved	None
4	2	2.50	1	1	2.50	1	Achieved	None	None	None
5	4	4.00	2	1	0.50	0	Achieved	Achieved	Achieved	Achieved
6	1	2.00	2	0	0.00	0	Achieved	Achieved	Achieved	Achieved
7	30	0.30	3	150	0.30	0	None	None	None	Achieved
8	360	0.25	0	32	0.25	0	Achieved	Achieved	None	None
9	20	2.00	0	5	2.50	0	Achieved	Achieved	None	None
10	2920	0.25	5	2700	0.25	2	Achieved	None	None	Achieved
11	28	1.00	3	0	0.00	0	Achieved	Achieved	Achieved	Achieved
12	4	20.00	1	0	0.00	0	Achieved	Achieved	Achieved	Achieved
13	12	5.00	3	0	0.00	0	Achieved	Achieved	Achieved	Achieved
14	5	3.00	2	0	0.00	0	Achieved	Achieved	Achieved	Achieved
15	16	15.00	2	26	5.00	0	None	None	Achieved	Achieved
16	12	4.00	2	2	2.00	0	Achieved	Achieved	Achieved	Achieved
17	8	2.50	1	46	0.50	0	None	None	Achieved	Achieved
18	4	0.20	1	6	0.20	1	None	None	None	None
19	32	2.00	0	3	2.00	0	Achieved	Achieved	None	None
20	1	0.50	0	0	0.00	0	Achieved	Achieved	Achieved	None
21	8	0.17	0	1	0.20	0	Achieved	Achieved	None	None
22	12	2.00	0	2	1.00	0	Achieved	Achieved	Achieved	None
23	24	10.00	3	114	5.00	2	None	None	Achieved	Achieved
24	4	4.00	3	0	0.00	0	Achieved	Achieved	Achieved	Achieved
25	12	10.00	2	6	1.00	0	Achieved	None	Achieved	Achieved
26	5	2.30	2	0	0.00	0	Achieved	Achieved	Achieved	Achieved
27	2	3.00	1	0	0.00	0	Achieved	Achieved	Achieved	Achieved
28	12	2.00	5	0	0.00	0	Achieved	Achieved	Achieved	Achieved
29	6	3.00	1	0	0.00	0	Achieved	Achieved	Achieved	Achieved
30	2	2.50	1	1	0.50	1	Achieved	None	Achieved	None
31	3	1.00	1	0	0.00	0	Achieved	Achieved	Achieved	Achieved

Note: Variables definition: Seizure severity is the cumulative score assessed by the shortened Ictal/Post‐ictal subscale of the Liverpool Seizure Severity Scale including smacking of lips or fidgeting, falling to the ground, headache and sleepiness following recovery, incontinence, tongue biting and injury during an attack. Variables units: Seizure duration is the average duration time to each seizure during the follow‐up period, with the unit of minutes; Unit of seizure frequency is the cumulative frequency for 1 year before or after surgery.

To investigate the effect of clinical information at baseline on the individual's seizure remission, we compared various variables based on whether the individual experienced a reduction in seizure frequency, average seizure duration and seizure severity, as details shown in Table 5. No clinical characteristics showed significant differences at the significant level of 0.05 between the group in remission and the group not in remission, but given that this case series report is a small sample study, some clinical characteristics that were significant at the significant level of 0.1 may be statistically significant in future studies with larger samples. Specifically, for the level of statistical significance at 0.1, there was a significant difference in the duration of ASMs taken and the presence of seizure aura between the remission group and that did not. The average duration of ASMs taken in the remission group was higher than in the non‐remission group (Remission group: 5.45 ± 5.49, Non‐remission group: 1.68 ± 2.48, p value of Wilcoxon test = 0.075). Additionally, the percentage of patients experiencing aura before a seizure in the remission group was lower than in the non‐remission group (remission group: 15.39%, non‐remission group: 60.00%, p value of Chi‐square test = 0.062).

TABLE 5.

The comparison of clinical information for epilepsy patients with and without remission.

Variables	Total	Remission frequency			Remission frequency over 50%			Remission duration			Remission severity
		Achieved, N = 26	None, N = 5	p Value	Achieved, N = 22	None, N = 9	p Value	Achieved, N = 22	None, N = 9	p Value	Achieved, N = 20	None, N = 11	p Value
Age	23.74 ± 10.70	23.96 ± 11.53	22.60 ± 5.03	0.809	23.82 ± 11.29	23.56 ± 9.72	0.794	22.64 ± 11.35	26.44 ± 8.89	0.276	23.00 ± 9.53	25.09 ± 12.95	0.999
Onset age	16.26 ± 11.86	16.04 ± 12.56	17.40 ± 8.20	0.666	16.50 ± 13.43	15.67 ± 7.31	0.983	15.18 ± 13.47	18.89 ± 6.37	0.077	14.65 ± 11.43	19.18 ± 12.62	0.229
Epilepsy course	7.71 ± 6.51	8.12 ± 6.85	5.60 ± 4.16	0.626	7.45 ± 6.43	8.33 ± 7.04	0.793	7.59 ± 6.43	8.00 ± 7.09	0.999	8.55 ± 6.74	6.18 ± 6.06	0.261
ASMs duration	4.84 ± 5.29	5.45 ± 5.49	1.68 ± 2.48	0.075	5.12 ± 5.25	4.15 ± 5.62	0.359	4.43 ± 5.22	5.83 ± 5.62	0.444	5.30 ± 5.89	4.01 ± 4.09	0.562
BMI	21.53 ± 3.41	21.68 ± 3.69	20.78 ± 1.25	0.830	20.80 ± 2.82	23.33 ± 4.21	0.128	21.02 ± 2.91	22.80 ± 4.35	0.361	21.00 ± 2.40	22.51 ± 4.73	0.620
Sex, (%)
Female	12 (38.71)	10 (38.46)	2 (40.00)	0.999	9 (40.91)	3 (33.33)	0.999	10 (45.45)	2 (22.22)	0.418	7 (35.00)	5 (45.45)	0.705
Male	19 (61.29)	16 (61.54)	3 (60.00)		13 (59.09)	6 (66.67)		12 (54.55)	7 (77.78)		13 (65.00)	6 (54.55)
ASMs types
≤1 type	16 (51.61)	12 (46.15)	4 (80.00)	0.333	10 (45.45)	6 (66.67)	0.433	13 (59.09)	3 (33.33)	0.252	9 (45.00)	7 (63.64)	0.537
>1 type	15 (48.39)	14 (53.85)	1 (20.00)		12 (54.55)	3 (33.33)		9 (40.91)	6 (66.67)		11 (55.00)	4 (36.36)
Etiological causes
Others	9 (29.03)	7 (26.92)	2 (40.00)	0.613	6 (27.27)	3 (33.33)	0.999	6 (27.27)	3 (33.333)	0.999	6 (30.00)	3 (27.27)	0.999
Unknown	22 (70.97)	19 (73.08)	3 (60.00)		16 (72.73)	6 (66.67)		16 (72.73)	6 (66.667)		14 (70.00)	8 (72.73)
Seizure types, (%)
Focal	18 (58.06)	14 (53.85)	4 (80.00)	0.805	10 (45.45)	8 (88.89)	0.092	12 (54.55)	6 (66.67)	0.999	11 (55.00)	7 (63.64)	0.891
Generalized	5 (16.13)	5 (19.23)	0 (0.00)		5 (22.73)	0 (0.00)		4 (18.18)	1 (11.11)		4 (20.00)	1 (9.09)
Unknown	8 (25.81)	7 (26.92)	1 (20.00)		7 (31.82)	1 (11.11)		6 (27.27)	2 (22.22)		5 (25.00)	3 (27.27)
Aura, (%)
None	24 (77.42)	22 (84.61)	2 (40.00)	0.062	18 (81.82)	6 (66.67)	0.384	16 (72.73)	8 (88.89)	0.639	14 (70.00)	10 (90.91)	0.372
Yes	7 (22.58)	4 (15.39)	3 (60.00)		4 (18.18)	3 (33.33)		6 (27.27)	1 (11.11)		6 (30.00)	1 (9.09)
Migraine, (%)
None	23 (74.19)	20 (76.92)	3 (60.00)	0.583	18 (81.82)	5 (55.56)	0.185	16 (72.73)	7 (77.78)	0.999	14 (70.00)	9 (81.82)	0.676
Yes	8 (25.81)	6 (23.08)	2 (40.00)		4 (18.18)	4 (44.44)		6 (27.27)	2 (22.22)		6 (30.00)	2 (18.18)
MRI, (%)
Abnormal	8 (25.81)	7 (26.92)	1 (20.00)	0.999	5 (22.73)	3 (33.33)	0.660	5 (22.73)	3 (33.33)	0.660	5 (25.00)	3 (27.27)	0.999
Normal	23 (74.19)	19 (73.08)	4 (80.00)		17 (77.27)	6 (66.67)		17 (77.27)	6 (66.67)		15 (75.00)	8 (72.73)
EEG, (%)
Abnormal	13 (41.93)	11 (42.31)	2 (40.00)	0.999	9 (40.91)	4 (44.44)	0.999	9 (40.91)	4 (44.44)	0.999	9 (45.00)	4 (36.36)	0.718
Normal	18 (58.07)	15 (57.69)	3 (60.00)		13 (59.09)	5 (55.56)		13 (59.09)	5 (55.56)		11 (55.00)	7 (63.64)

Note: Variables definition: Sex including female and male; ASMs types represents the number of ASM categories taken by patients in the week before surgery, including Valproate (VPA), Levetiracetam (LEV), Lamotrigine (LTG), Topiramate (TPM), Oxcarbazepine (OXC), Phenobarbital (PB), Carbamazepine (CBZ); ASMs duration represents the length of time from the time at which the patients started the antiseizure medications until 1 week before surgery; etiological causes including structural, genetic, infectious, metabolic, immune and unknown; seizure types including focal onset, generalized onset and unknown onset; aura and migraine including No and Yes; abnormal EEG including sharp, spike and slow waves found in the frontal, parietal, temporal, and occipital lobes of the brain and in the hippocampus or globally; Abnormal MRI including obvious bleeding foci, arachnoid cysts or hippocampal sclerosis were found in the frontal, parietal, temporal and occipital lobes of the brain and the hippocampus. Variables units: units of age, onset age, epilepsy course and ASMs duration are years; unit of BMI is kg/m². Variables description: Age, BMI, onset age, epilepsy course, and ASMs duration are described by mean value plus standard deviation; sex, ASMs types, etiological causes, seizure types, aura, migraine, EGG and MRI are described by frequency plus percentage; others group in the etiological causes including structural, genetic and infectious etiologies.

Abbreviations: ASMs, antiseizure medications; BMI, body mass index; EEG, electroencephalography; MRI, magnetic resonance imaging.

When considering whether the reduction in seizure frequency is >50%, only the types of seizure exhibited a significant contrast between the two groups. The percentage of focal seizures in the remission over 50% group is significantly lower than those in the compared group (remission over 50% group: 45.45%, non‐remission over 50% group: 88.89%, p value of Chi‐square test = 0.092). The comparison of secondary outcomes revealed a significant difference in the onset age of seizure between groups based on whether patients achieved remission of seizure duration (remission group: 15.18 ± 13.47, non‐remission group: 18.89 ± 6.37, p value of Wilcoxon test = 0.077). And there was no significant difference in clinical information between the group that achieved remission of seizure severity and that did not.

3.3. Safety of PFO closure

On the initial day after surgery, chest X‐rays and TTE confirmed that all cardiac occluders were properly positioned, and no complications led to extended hospital stay. During a follow‐up period of at least 1 year, no patients had a thromboembolic event or new bleeding, and there were no postoperative issues such as paroxysmal supraventricular tachycardia, atrial fibrillation, etc. Only one patient complained about chest stinging pain in 3rd month, however, the Electrocardiograph (ECG) revealed sinus tachycardia and a TTE test showed that the occluder was in the correct position and the patient recovered spontaneously after a few days. Antiplatelet therapy was withdrawn in one case in the 3rd month because of the presence of melena which was determined to be caused by gastrointestinal bleeding as detected with a gastroenteroscope. However, all of the other patients were maintained on the antiplatelet regimen above.

4. DISCUSSION

Based on our long‐term monitoring of patients with drug‐resistant epilepsy and severe RLS, we found a significant reduction in the frequency, average duration and severity of seizures after 1 year of surgery. This finding may contributed to a novel treatment for those epilepsy patients.

While PFO is a common abnormality in most individuals, it allows for the passage of paradoxical embolus, molecule, and partially deoxygenated blood from the right side of the heart to the left side and the brain, bypassing pulmonary circulation. Therefore, there is great need for a safe and straightforward interventional procedure, such as percutaneous closure of the PFO, in various conditions including stroke, migraine, obstructive sleep apnea (OSA), and decompression sickness. ²⁷ , ²⁸ However, the safety and effectiveness of PFO closure in epilepsy patients remain uncertain. Our findings indicate that epilepsy patients with PFO benefited from the procedure of PFO closure as it led to a reduction in seizures frequency and duration, as well as improvement in seizure severity.

In addition, this study's findings, which categorized patients based on whether they achieved seizure remission and compared their baseline clinical information, provided valuable insights. First, taking ASMs for a longer period was associated with better remission in seizure frequency. Additionally, seizures without aura, generalized seizures, and seizures of unknown origin were also associated with reduction of seizure frequency. Second, an earlier onset age was associated with shorter seizure duration. These clinical phenomena all suggested that PFO closure may offer a more direct and efficient treatment for epilepsy patients of this type. This hypothesis should be confirmed through larger clinical studies and explored further through mechanism studies. In terms of safety, no serious adverse events were reported during PFO closure procedures, and all patients achieved successful plugging effect, demonstrating the safety of this approach.

The significant effect of PFO closure on controlling seizures suggested that there might exist an association between changes in blood substances and epileptic seizure, and offering long‐term benefits for epilepsy patients. One potential explanation is that the presence of PFO leads to a reduction in an individual's blood oxygen saturation, which could potentially contribute to the development of hypoxia in the brain. ²⁹ And some research has reported that prolonged exposure to hypoxia might induce self‐generated spiking activity and increased susceptibility to seizures. ³⁰ Of note, our findings that patients who had a remission of seizures took a longer average duration of ASMs than did patients who did not have a remission, which might be associated with the mechanism of hypoxia induced seizures. In a mechanistic study by Dong et al., they found that the postoperative blood oxygen partial pressure and oxygen saturation of patients with migraine and epilepsy were significantly elevated after PFO closure. Furthermore, multisite (peripheral, right and left atrial) and multitimepoint (before and after surgery) plasma proteomics from patients showed that the levels of free hemoglobin and cell adhesion molecules (CAMs) were significantly increased after PFO closure, which may be related to the relief of hypoxic state. ³¹ Therefore, we speculated that if the patients could maintain long‐term desired ASMs management before surgery, it can continuously and stably control seizures to a certain extent, and improved the symptoms of hypoxia caused by seizures; so that the improvement of hypoxia after surgery may have a better effect on the remission of seizures. However, since this is only a case series report of 31 patients, so this speculation needs to be verified by clinical trials and basic research with larger sample sizes in the future.

On the other side, the large RLS volume caused by PFO could lead air and micro‐thrombus to bypass the pulmonary filter and enter the blood vessel of brain, which could trigger cortical spreading depression (CSD). Some studies have reported that CSD could heighten neuronal excitability and trigger epileptiform discharges in partially disinhibited neuronal tissues, which has shown its role in post‐stroke epilepsy. ³² Besides, CSD also could activate neuronal Panx1 channels, contributing to neuronal hyperactivity during seizures. ³³ Furthermore, the recurrence rate and amplitude of spontaneous epileptiform abnormal discharges have been found to be increased by CSD, both in vivo and in human hippocampal slices. ³⁴ Given these evidence, PFO‐triggered CSD might be an important mechanism to inducing the epileptic seizures, but further mechanism studies is needed to provide conclusive evidence in the future.

Here were some strengths and limitations in this prospective observational case series study. As far as we know, it was the first time that the postoperative seizure remission in epileptic patients who underwent PFO closure was observed, providing critical clinical evidence for treating epileptic seizures with PFO closure. This study also revealed a new disease association for the heart–brain axis. However, there still existed some limitations, first, the number of cases we observed was still very limited, which restricted the statistical power of the analysis. Second, we did not establish stringent criteria for including or excluding epilepsy patients, especially some patients took ASMs irregularly or for not long enough before surgery, and some patients had clear etiological causes, which may have a certain impact on the results of this study; and the postoperative‐modified ASMs regimens for several patients could also cause bias to a certain extent. In addition, patients were limited by their own economic conditions or educational level, and their compliance with ASMs was not ideal. Many patients may chose to reduce or even not take ASMs drugs when various ASMs drugs were ineffective, which may interfere with the evaluation of surgical efficacy. Besides, although it does not affect our main conclusion, a few patients had not yet been diagnosed with drug‐resistant epilepsy, which may cause a bias in the evaluation of therapeutic efficacy for PFO closure. Furthermore, our study was not a randomized controlled trial, which makes it possible that unaccounted confounding factors might potentially impact the therapeutic effectiveness of PFO closure. Finally, there have been indications from other studies that aspirin might have a potential effect on controlling seizures, and our study cannot dismiss the potential influence of this antiplatelet medication.

5. CONCLUSION

The PFO closure has been shown for the first time to result in a significant remission in the frequency, duration and severity of seizures. Patients with drug‐resistant epilepsy and high‐grade RLS of PFO are the appropriate candidates for PFO closure, which has been identified to have a favorable safety profile. Further validation of these findings is necessary through more rigorous randomized controlled trials and studies exploring the underlying mechanisms.

AUTHOR CONTRIBUTIONS

LC designed the study. SJ, BD, YT, HL, WL, YL, YC and AP retrieved, collected, anonymized, curated, and verified the data. SJ and BD analyzed the data, interpreted the results, and produced the figures. The manuscript was written by all authors and reviewed by LC and SJ. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.

FUNDING INFORMATION

We acknowledge the support from the Chinese Academy of Medical Sciences by Medical and Health Technology Innovation Project (No. 2022‐I2M‐C&T‐B‐100).

CONFLICT OF INTEREST STATEMENT

None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

ETHICS STATEMENT

This study had received ethical approval from the Ethics Committee of West China Hospital of Sichuan University, and registered on the Chinese Clinical Trial Registry (Registration number: ChiCTR‐OOC‐17011935). All participants were required to sign the informed consent.

Supporting information

Appendix S1.

Ji S, Dong B, Tang Y, Li H, Lai W, Li Y, et al. Therapeutic value of patent foramen ovale closure for drug‐resistant epilepsy: A case series report. Epilepsia Open. 2024;9:1357–1371. 10.1002/epi4.12960

DATA AVAILABILITY STATEMENT

Due to the nature of this research using data from the hospital's case management system, participants of this study did not agree for their data to be shared publicly, so supporting data are not available.